The Bishop
Labour
Down With Factionalism!
Posts: 36,483
|
Post by The Bishop on Oct 17, 2021 10:48:27 GMT
Why changes from July and not September? Possibly differing methodologies - I recall when ComRes used to regularly do both online and phone surveys - but assuming that is not the case here it will likely be the polls being published in different media outlets. Which is rather more annoying.
|
|
|
Post by Davıd Boothroyd on Oct 17, 2021 10:55:23 GMT
It may be that the July one was the most recent to have been commissioned by the Mail on Sunday, and the ones after that were just done as part of Deltapoll's routine polling.
There is a good case to be made for not looking at poll changes from one poll to the next, as they're far more affected by random sampling bounces, but instead looking at changes over a longer period.
|
|
|
Deltapoll
Oct 17, 2021 12:59:20 GMT
via mobile
Post by andrew111 on Oct 17, 2021 12:59:20 GMT
It may be that the July one was the most recent to have been commissioned by the Mail on Sunday, and the ones after that were just done as part of Deltapoll's routine polling. There is a good case to be made for not looking at poll changes from one poll to the next, as they're far more affected by random sampling bounces, but instead looking at changes over a longer period. Yes, but comparing with a single poll from longer ago is really daft! However if Delta want to compare all future polls with the one where the Lib Dems got 6%, I say "fair play, well justified analysis"
|
|
johng
Labour
Posts: 4,491
|
Post by johng on Nov 6, 2021 14:17:09 GMT
This surely has to be an outlier? Even with most of the fieldwork done before Paterson's resignation.
|
|
jamie
Top Poster
Posts: 6,839
|
Post by jamie on Nov 6, 2021 15:00:13 GMT
With Others on 15% they really ought to let us know how that figure is split between the minor parties. Green - 6% SNP - 3% REFUK - 2% UKIP - 2% PC - 1% Others - 1% It would seem sensible to include the Greens in the initial announcement given 6% is not much different from the Lib Dems 8%, and I'm not sure the continued relevance of prompting for UKIP (Deltapoll were the only pollster from early November 2019 onwards to have UKIP above 0%, they got 0.1%).
|
|
Sibboleth
Labour
'Sit on my finger, sing in my ear, O littleblood.'
Posts: 15,265
|
Post by Sibboleth on Nov 6, 2021 15:59:06 GMT
This surely has to be an outlier? Even with most of the fieldwork done before Paterson's resignation. Not necessarily, the thing to note is that 2pts is well within the MoE. A lot of poll movement (in all directions) is statistical noise.
|
|
cogload
Lib Dem
I jumped in the river and what did I see...
Posts: 8,022
|
Post by cogload on Dec 4, 2021 20:13:17 GMT
|
|
|
Post by yellowperil on Dec 4, 2021 20:28:51 GMT
Do we have a breakdown of the "others"?
|
|
|
Post by lackeroftalent on Dec 4, 2021 20:43:45 GMT
|
|
|
Post by Davıd Boothroyd on Jan 1, 2022 21:18:51 GMT
I presume this is a DeltaPoll - they are the usual pollsters for the MoS.
|
|
|
Post by woollyliberal on Jan 1, 2022 22:11:25 GMT
Yes, it is Deltapoll
national (red wall) Con 35 (33) Lab 40 (49) LD 10 (8)
/photo/1
|
|
jamie
Top Poster
Posts: 6,839
|
Post by jamie on Jan 2, 2022 0:10:20 GMT
Yes, it is Deltapoll national ( red wall) Con 35 (33) Lab 40 (49) LD 10 (8) The Mail misreported this part, it’s actually 2019 Conservative gains, so quite a few seats that cannot be included under the term ‘red wall’ (not that any seat should be of course).
|
|
|
Post by woollyliberal on Jan 2, 2022 8:58:41 GMT
Yes, it is Deltapoll national ( red wall) Con 35 (33) Lab 40 (49) LD 10 (8) The Mail misreported this part, it’s actually 2019 Conservative gains, so quite a few seats that cannot be included under the term ‘red wall’ (not that any seat should be of course). That is a better poll but it means something worse for the new Tory intake. All those MPs who won from Labour in 2019 should be planning for life after the next election. The words "Boris Johnson is a drag on the ticket" may ring in the ears of those who might try to hold their seat. It may be too early for emails to Graham Brady yet/, but sounding out the best replacement will be high on the agenda now.
|
|
clyde1998
SNP
Green (E&W) member; SNP supporter
Posts: 1,765
|
Post by clyde1998 on Jan 2, 2022 9:32:50 GMT
It appears the Conservative gains section is a 'boosted' sub-sample: deltapoll.co.uk/polls/voteint220101. For completeness: 2019 Conservative gains Lab 49% (+12) Con 33% (-15) LDm 8% (+1) Grn 4% (+3) RUK 3% (-2) PC 2% (+1) UKIP 2% (+2) Oth 0% (-1)2019 Conservative holdsCon 47% (-11) Lab 26% (+3) LDm 14% (nc) Grn 6% (+3) RUK 6% (+6) UKIP 2% (+2) SNP 0% (-1) PC 0% (-1) Oth 1% (nc)2019 seats not won by ConservativesLab 46% (+1) Con 26% (-1) SNP 13% (+4) LDm 10% (+1) RUK 2% (-2) Grn 2% (-1) PC 1% (nc) UKIP 1% (+1) Oth 0% (-1)Each group is around a third of the total sample, once weighted. Seems to be very little change, on the surface, outside of Conservative-held seats (I'd ignore the SNP & PC figures). In the seats the Conservatives gained, Labour are the biggest beneficiaries of the collapse in support - although almost all Conservative gains were from Labour. Reform seem to be the biggest winners in Conservative holds, no doubt partly due to not having candidates in 2019. I wonder if a lot would return to the Conservatives if an election was held now.
|
|
|
Deltapoll
Jan 2, 2022 10:48:41 GMT
via mobile
Post by andrew111 on Jan 2, 2022 10:48:41 GMT
It appears the Conservative gains section is a 'boosted' sub-sample: deltapoll.co.uk/polls/voteint220101. For completeness: 2019 Conservative gains Lab 49% (+12) Con 33% (-15) LDm 8% (+1) Grn 4% (+3) RUK 3% (-2) PC 2% (+1) UKIP 2% (+2) Oth 0% (-1)2019 Conservative holdsCon 47% (-11) Lab 26% (+3) LDm 14% (nc) Grn 6% (+3) RUK 6% (+6) UKIP 2% (+2) SNP 0% (-1) PC 0% (-1) Oth 1% (nc)2019 seats not won by ConservativesLab 46% (+1) Con 26% (-1) SNP 13% (+4) LDm 10% (+1) RUK 2% (-2) Grn 2% (-1) PC 1% (nc) UKIP 1% (+1) Oth 0% (-1)Each group is around a third of the total sample, once weighted. Seems to be very little change, on the surface, outside of Conservative-held seats (I'd ignore the SNP & PC figures). In the seats the Conservatives gained, Labour are the biggest beneficiaries of the collapse in support - although almost all Conservative gains were from Labour. Reform seem to be the biggest winners in Conservative holds, no doubt partly due to not having candidates in 2019. I wonder if a lot would return to the Conservatives if an election was held now. 2% for UKIP too. The other way of looking at is to say that REFUK standing will hurt the Tories by up to 6-8% in seats held last time.. In that category the Lib Dems woukd make about 10 gains and SNP too, I guess. Labour quite a few especially if the Greens are squeezed as usual, which would make up for a partial return of REFUK voters The tiny swing in non-Con seats suggests dodgy subsamples that overestimate what is happening in the other two categories
|
|
The Bishop
Labour
Down With Factionalism!
Posts: 36,483
|
Post by The Bishop on Jan 2, 2022 11:31:35 GMT
The tiny swing in non-Con seats suggests dodgy subsamples that overestimate what is happening in the other two categories Whilst that does appear to be a low swing, it is certainly believable that Tory holds/gains in 2019 have moved against the party more.
|
|
|
Deltapoll
Jan 2, 2022 12:01:44 GMT
via mobile
Post by andrew111 on Jan 2, 2022 12:01:44 GMT
The tiny swing in non-Con seats suggests dodgy subsamples that overestimate what is happening in the other two categories Whilst that does appear to be a low swing, it is certainly believable that Tory holds/gains in 2019 have moved against the party more. I agree, but the Tories should still be losing more than 1% everywhere. Take Huddersfield, the Tory vote rose by 10% from 2015 to 2021, within 2% of the national vote change.
|
|
The Bishop
Labour
Down With Factionalism!
Posts: 36,483
|
Post by The Bishop on Jan 2, 2022 12:09:50 GMT
Maybe it actually suggests that Labour "should" be more than 5% ahead nationally.
|
|
|
Deltapoll
Jan 2, 2022 20:04:23 GMT
via mobile
Post by andrew111 on Jan 2, 2022 20:04:23 GMT
Maybe it actually suggests that Labour "should" be more than 5% ahead nationally. It might suggest that, and in most recent polls they have been. Dodgy subsamples do suggest an overall error, but the polls need a bit of time to settle now people have put 2021 behind them
|
|
|
Deltapoll
Jan 2, 2022 20:40:00 GMT
via mobile
Post by frankyank on Jan 2, 2022 20:40:00 GMT
What was the BXP % in those Con gains? It seems to have been reweighed or distributed to Lab/Con, judging from the changes posted.
|
|